Trying to make a shell script - need help

johnmcdnl

Hey lads... I'm just starting to use shell scripts but I'm far from great at it...

I'm basically trying to write a script to take 3-4 files and compare the length of them and then output the name of the longest file...

eg.. essay.txt music.txt document.txt

then have the output as essay.txt as the essay is the longest file..

I know I need to do some form of arguments such as $1 $2 and $3 but I'm really stuck at how to start these scripts...

any help would be really appreciated lads..

Thanks

Skrynesaver

ls -l | awk '{print $5 " " $9}'|sort -n| tail -1 |cut -d\  -f2

ls -l # get the full listing of the files
awk '{print $5 " " $9}' # extract the size and name
sort -n # sort by size
tail -1 # Take the last entry (largest file)
cut -d\ -f2 # drop the size and print the filename

There is one obvious problem here, if you don't shoot users who put spaces in filenames the displayed filename will be truncated at the first space.

If you need to use this in a shell script that takes arguments try the following

ls -l $* | awk '{print $5 " " $9}'|sort -n| tail -1 |cut -d\  -f2

pickarooney

There may be a slight difference in the result depending on whether you want the file with most words/lines or most bytes (e.g. a 500 line poem might be longer than a 400 line essay but have fewer words). If this is the case, maybe use wc -l instead of sort.

Also, are all the files, and only those files, in the same directory or do you want to be able to specify the files to compare at runtime?

johnmcdnl

when i'm running the script i just want to say.. sh script.sh /home/username/essay.txt /home/username/poem.txt etc tc

I got it working anyways so thanks lads

johnmcdnl

I'm not gonna make a new thread so I'm trying something new

I want to write a sentence out and do some kind of command to find the longest word in the sentence...

So say - script.sh My name is johnmcdnl

will output "johnmcdnl"

any ideas - once again thanks in advance guys

Skrynesaver

#!/bin/sh
perl -e '$lenMax=0;for(@ARGV){if (length($_) > $lenMax){$string = $_ ;$lenMax=length($string);}}print "$string\n";' $*

Sorry, but that's a bit much for a shell without creating a load of temp files etc...

johnmcdnl

Skrynesaver wrote: »

#!/bin/sh
perl -e '$lenMax=0;for(@ARGV){if (length($_) > $lenMax){$string = $_ ;$lenMax=length($string);}}print "$string\n";' $*

Sorry, but that's a bit much for a shell without creating a load of temp files etc...

cheer mate :cool:

I'm assuming that's the easiest way to do it.. I though there would be an easier way to do it but thanks anyways...

Skrynesaver

johnmcdnl wrote: »

cheer mate :cool:

I'm assuming that's the easiest way to do it.. I though there would be an easier way to do it but thanks anyways...

'twas a sort of a joke, the more legible way to do it would be

#!/usr/bin/perl
use strict;
use warnings;
my $long_word="";
my $lenMax=0;
for my $word (@ARGV){
   if (length($word) > $lenMax){
      $long_word = $word ;
      $lenMax=length($string);
   }
}
print "$long_word\n";

Snowbat

The original problem...

Skrynesaver wrote: »

There is one obvious problem here, if you don't shoot users who put spaces in filenames the displayed filename will be truncated at the first space.

I'd rather shoot the coder who didn't allow for perfectly valid filenames
How about

ls -S1 | head -n 1

or

ls -S1 'essay.txt' 'music.txt' 'filename with space.txt' | head -n 1

Second problem...

for word in 'one' 'thirty three' 'two' 'three' 'fourty four'; do echo `echo $word | wc -m` $word; done | sort -nr | head -n 1 | cut -d ' ' -f1 --complement

for word in My name is johnmcdnl; do echo `echo $word | wc -m` $word; done | sort -nr | head -n 1 | cut -d ' ' -f1 --complement

@johnmcdnl Do you need these in script form?